eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
eroux has joined #haiku
euandreh has joined #haiku
<kallisti5[m]>
if anyone is looking for a project, you should fork https://github.com/usefathom/fathom and maintain it :-) It's go-lang and we use it for metrics.haiku-os.org
vdamewood has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
pch_ is now known as kinkinkijkin
<kinkinkijkin>
I'm talking with another person involved with cadmium-linux (an installer for installing non-google linux on arm chromebooks) about something pertaining to the boot process, and I've realized it might be possible to hook the firmware menu somehow to create an EL2 passoff for writing more standard bootloading processes in
<kinkinkijkin>
I then realized if this is possible it would mean that it could be possible to leverage the standardized platform of arm chromebooks for early, cheap hardware testing of haiku for arm64
<kinkinkijkin>
whereas a similarly standard set of boards would otherwise run thousands of dollars each or require making rpi-isms
<kinkinkijkin>
I'm not interested in developing such a thing myself but it might possibly be an idea for the arm64 maintainers to consider and possibly contribute to a bounty for
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
eroux has joined #haiku
eroux has quit [Remote host closed the connection]
<nekobot2>
[haiku/haiku] 1da90c98f23a - Enable and fix WError in add-ons/kernel/busses/
m199 has joined #haiku
<selfish>
Hello, seeing some confusing behaviour from poll/select on sockets
<selfish>
Seems poll/select reports listening sockets as "outputable"; apparently because socket_request_notification notifies when read_avail() < B_OK; TCPEndpoint::SendAvailable (called from the read_avail for TCP if I'm reading right) seems to return a (0 or positive) buffer size for writable sockets or EPIPE otherwise
<selfish>
Am I missing something, or are the socket_request_notification checks inverted (should notify when available > B_OK)?
<selfish>
(in the source tree, socket_request_notification on src/add-ons/kernel/network/stack/net_socket.cpp and SendAvailable on ../protocols/tcp/TCPEndpoint.cpp from the previous folder)
Atomozero has joined #haiku
jmairboeck has quit [Quit: Konversation terminated!]
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<nephele>
iirc error codes are all negative int32 do < B_OK would mean no error
<nephele>
i think checking for == B_OK or != B_OK is much clearer when reqding code though
Atomozero has quit [Quit: Vision[]: i've been blurred!]
<selfish>
> B_OK would be no error (B_OK is 0; x > B_OK tests for positive x), no? It can't check for == B_OK since socket_read_avail might return a positive integer (the TCP function returns the free buffer size), which is what it seems to want to test for, if I'm reading correctly
x10z has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
doomquak1keen has quit [Ping timeout: 480 seconds]